A vehicle adaptive braking optimization method and device for a complex scene and a medium

Abstract

The application discloses a kind of complex scene's vehicle adaptive braking optimization method, device and medium, the method includes: in the case where pedestrian is in non-crossing scene, increase the bias value threshold of pedestrian intrusion vehicle driving path, obtain non-crossing scene adjustment result;In the case where pedestrian is in crossing scene, filter the lateral velocity of pedestrian, obtain crossing scene adjustment result;In the case where pedestrian is in curve scene, the curvature value of vehicle trajectory is reduced to preset value, obtain curve scene adjustment result.The application proposes a kind of complex scene's vehicle adaptive braking optimization method, device and medium, the adaptive adjustment of vehicle braking mode under non-crossing scene, crossing scene adjustment and curve scene is carried out, jointly constitute the emergency braking adjustment result of vehicle, can fully consider actual situation and potential risk under different scenes, can solve the problem that vehicle is prone to false triggering collision warning and automatic emergency braking in complex scene.

Description

TECHNICAL FIELD

[0001] The present application relates to automatic driving technology, in particular to a vehicle adaptive braking optimization method and device for complex scenes and a medium. BACKGROUND

[0002] The vehicle model equipped with a pure driving assistance camera, a driving assistance camera + radar or a radar can detect front vehicles (motor vehicles & non-motor vehicles) or pedestrians in real time, judge the distance, direction and relative speed between the vehicle and the front vehicle (pedestrian), and guide the driver to handle in time through voice prompt when there is potential collision danger and the distance from the front vehicle is less than the safety range. The ADAS vision subsystem provided will identify pedestrians around the vehicle, detect the distance between the pedestrians and the vehicle, and make a warning decision accordingly, thereby providing assistance for safe driving. The traditional pure driving assistance camera scheme mainly uses a camera as a sensor to detect the distance between pedestrians and vehicles through image processing algorithms, has the advantages of low cost, easy installation and easy integration into existing vehicle systems, and plays an important role in detecting the distance between pedestrians and vehicles.

[0003] However, in the actual debugging or road test process, the pure driving assistance camera scheme is limited to the inaccuracy of the pure visual scheme in judging the target longitudinal distance and speed, especially for complex road conditions, especially pedestrians. It is easy to misselect targets in this complex scene, and it is also easy to mistrigger FCW false alarms or AEB functions before reaching the trigger emergency scene, which interferes with the normal driving of the driver and may even cause rear-end collisions. SUMMARY

[0004] The present application provides a vehicle adaptive braking optimization method, device and medium for complex scenes to solve the problem of easy mis-triggering of collision warning and automatic emergency braking of vehicles in complex scenes.

[0005] In the case that the pedestrian is in a non-crossing scene, a bias value threshold of the pedestrian invading the driving path of the vehicle is increased to obtain a non-crossing scene adjustment result;

[0006] In the case that the pedestrian is in a crossing scene, the lateral speed of the pedestrian is filtered to obtain a crossing scene adjustment result;

[0007] In the case that the pedestrian is in a curve scene, the curvature value of the vehicle trajectory is reduced to a preset value to obtain a curve scene adjustment result;

[0008] The non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result constitute an emergency braking adjustment result of the vehicle.

[0009] In the present application, when the pedestrian is in a non-crossing scene, if the distance between the pedestrian and the vehicle is too close but there is no actual collision risk, by increasing the bias value threshold of the pedestrian invading the driving path of the vehicle, the AEB system of the vehicle can more flexibly consider the relative position and distance between the pedestrian and the vehicle when judging whether to trigger emergency braking. In the pedestrian crossing scene, by filtering the lateral speed information of the pedestrian, the actual motion state of the pedestrian can be more accurately identified, and when the pedestrian does indeed have the risk of crossing the driving path of the vehicle, the AEB system will trigger emergency braking; when the pedestrian is only temporarily staying or walking slowly, the system will not be triggered. In the curve scene, due to the change of the driving trajectory and the front view of the vehicle, the traditional AEB system may misjudge the object inside the curve or the obstacle on the roadside in front of the vehicle as a potential collision risk, thereby triggering emergency braking; by reducing the curvature value of the vehicle trajectory to a preset value, the AEB system can be more flexible and accurate when judging the curve scene.

[0010] Compared with the prior art, the present application adjusts the vehicle braking mode in non-crossing scenes, crossing scenes and curve scenes, which together constitute the emergency braking adjustment result of the vehicle. This comprehensive adjustment result can fully consider the actual situation and potential risks in different scenes, making the AEB system more accurate and reliable when judging whether to trigger emergency braking, thereby solving the problem of false triggering of collision warning and automatic emergency braking of the vehicle in complex scenes.

[0011] As a preferred scheme, in the case that the pedestrian is in a non-crossing scene, the bias value threshold of the pedestrian invading the driving path of the vehicle is increased to obtain a non-crossing scene adjustment result, specifically:

[0012] In the case that the pedestrian is in a non-crossing scene, the bias value threshold of the vehicle driving path in the vehicle is adjusted in different dimensions to obtain a bias value threshold adjustment result;

[0013] The confirmation frame of the threshold time of the pedestrian is increased to obtain a confirmation frame adjustment result;

[0014] The non-crossing scene adjustment result is composed of the bias value threshold adjustment result and the confirmation frame adjustment result.

[0015] The preferred scheme can more accurately control the driving path of the vehicle by adjusting the bias value threshold in different dimensions, so that the vehicle can make more delicate and accurate avoidance actions when encountering pedestrians. Increasing the confirmation frame of the threshold time of the pedestrian can enable the vehicle to detect the presence of the pedestrian at an earlier time and make corresponding warnings and responses.

[0016] As a preferred solution, in the case that the pedestrian is in a non-crossing scene, the bias value threshold of the vehicle driving path in the vehicle is adjusted in different dimensions to obtain a bias value threshold adjustment result, specifically:

[0017] Based on the case that the pedestrian is in a non-crossing scene, in one-dimensional conditions, the lateral speed of the pedestrian is set to be less than a preset value; in two-dimensional conditions, the longitudinal speed of the pedestrian is set to be less than a preset value; in three-dimensional conditions, the vehicle speed of the vehicle is set to be less than a preset value, to obtain the bias value threshold adjustment result.

[0018] In one-dimensional conditions, by setting the lateral speed of the pedestrian to be less than a preset value, the preferred solution can accurately distinguish whether the pedestrian is in a state of stillness, slow walking, or preparing to cross the road, which helps the system to identify and react in advance, avoiding potential collision risks. In two-dimensional conditions, considering the longitudinal speed of the pedestrian, the behavior pattern of the pedestrian can be further refined, such as distinguishing whether the pedestrian is accelerating, decelerating, or walking at a constant speed, thereby more accurately judging its impact on the vehicle driving path. In three-dimensional conditions, by setting the vehicle speed of the vehicle to be less than a preset value, it can ensure that the vehicle maintains a safe speed when driving near the pedestrian, reducing the risk of collision due to excessive speed.

[0019] As a preferred solution, after obtaining the non-crossing scene adjustment result, it further includes:

[0020] According to a preset bias formula set, the closest angular distance and the farthest angular distance between the vehicle camera perception frame and the pedestrian are calculated; wherein the bias formula set includes a nearest distance calculation formula and a farthest distance calculation formula;

[0021] If the closest angular distance and the farthest angular distance do not meet the preset threshold, and the walking state of the pedestrian at the current time and the future predicted time does not meet the non-crossing scene adjustment result, the automatic emergency braking of the vehicle is triggered.

[0022] Through the preset bias formula set, the closest angular distance and the farthest angular distance between the vehicle camera perception frame and the pedestrian can be accurately calculated, which helps the system to grasp the position information of the pedestrian in real time, providing accurate basis for subsequent avoidance and braking decisions. Through the preset bias formula set and threshold judgment, the system can quickly make a decision on whether to trigger automatic emergency braking in a short time, and this efficient decision-making mechanism helps to reduce system response time and improve overall response efficiency.

[0023] As a preferred solution, the nearest distance calculation formula is specifically:

[0024] closeoffset = closeoffset1 + 0.5 * obj_width

[0025] wherein, closeoffset is the closest angular distance of the adjusted camera perception box to the pedestrian, closeoffset1 is the closest angular distance of the unadjusted camera perception box to the pedestrian, and obj_width is the pedestrian target width.

[0026] As a preferred solution, when the pedestrian is crossing the scene, the lateral speed of the pedestrian is filtered to obtain a crossing scene adjustment result, specifically:

[0027] If the longitudinal speed of the pedestrian is greater than the lateral speed of the pedestrian, or the lateral speed of the pedestrian is less than or equal to a preset value, the vehicle brake confidence of the vehicle is set to be untrusted to obtain a confidence setting result.

[0028] Under the condition of the confidence setting result, the pedestrian recognition time of the vehicle is set to obtain the crossing scene adjustment result.

[0029] In this preferred solution, when the longitudinal speed of the pedestrian is greater than the lateral speed of the pedestrian, it may mean that the pedestrian is straightening rather than crossing the road; at this time, if the system mistakenly identifies the pedestrian as crossing the scene and triggers emergency braking, it may cause unnecessary danger and accidents; therefore, setting the brake confidence to be untrusted can avoid such misjudgment, thereby improving driving safety. And if the lateral speed of the pedestrian is less than or equal to a preset value, it may indicate that the pedestrian is in a stationary or slow-moving state, in which case setting the brake confidence to be untrusted can reduce emergency braking triggered by slight movement of the pedestrian, avoiding unnecessary shock and discomfort to the driver and passengers.

[0030] As a preferred solution, after obtaining the curve scene adjustment result, it further includes:

[0031] If the speed of the current vehicle is less than or equal to a first preset value, the curvature value of the vehicle trajectory is calculated according to the steering wheel angle of the current vehicle;

[0032] If the speed of the current vehicle is greater than or equal to a second preset value, the curvature value of the vehicle trajectory is calculated according to the yaw rate of the current vehicle;

[0033] If the speed of the current vehicle is greater than the first preset value and less than the second preset value, the curvature value of the current vehicle trajectory is calculated according to the steering wheel angle and the yaw rate of the current vehicle;

[0034] If the curvature value of the current vehicle trajectory is greater than the preset value, the automatic emergency braking of the current vehicle is triggered.

[0035] In the preferred solution, when the vehicle is running at low speed, the change of the steering wheel angle has a greater impact on the vehicle trajectory, and therefore the trajectory curvature value calculated by the steering wheel angle can more accurately reflect the actual driving state of the vehicle. When the vehicle is running at high speed, the change of the yaw rate has a more significant impact on the vehicle trajectory; at this time, the trajectory curvature value calculated by the yaw rate can more accurately predict the driving trajectory of the vehicle, thereby improving the driving safety. When the vehicle is running at medium speed, the changes of the steering wheel angle and the yaw rate both have an impact on the vehicle trajectory; therefore, comprehensive consideration of both can more accurately calculate the trajectory curvature value, improving the comprehensiveness and accuracy of the calculation.

[0036] The application also provides a vehicle adaptive braking optimization device for a complex scene, comprising a non-crossing module, a crossing module, a curve module and a comprehensive module.

[0037] The non-crossing module is configured to increase the bias value threshold of the pedestrian invading the driving path of the vehicle in the case of a non-crossing scene to obtain a non-crossing scene adjustment result.

[0038] The crossing module is configured to filter the lateral speed of the pedestrian in the case of a crossing scene to obtain a crossing scene adjustment result.

[0039] The curve module is configured to reduce the curvature value of the vehicle trajectory to a preset value in the case of a curve scene to obtain a curve scene adjustment result.

[0040] The comprehensive module is configured to constitute the emergency braking adjustment result of the vehicle by the non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result.

[0041] As a preferred solution, the non-crossing module comprises a bias unit, a time unit and a combination unit.

[0042] The bias unit is configured to adjust the bias value threshold of the vehicle in the driving path in different dimensions in the case of a non-crossing scene to obtain a bias value threshold adjustment result.

[0043] The time unit is configured to increase the confirmation frame of the pedestrian existence threshold time to obtain a confirmation frame adjustment result.

[0044] The combination unit is configured to constitute the non-crossing scene adjustment result by the bias value threshold adjustment result and the confirmation frame adjustment result.

[0045] As a preferred solution, the bias unit specifically comprises:

[0046] In a one-dimensional condition, the lateral velocity of the pedestrian is set to be less than a preset value; in a two-dimensional condition, the longitudinal velocity of the pedestrian is set to be less than a preset value; and in a three-dimensional condition, the vehicle velocity of the vehicle is set to be less than a preset value, to obtain the bias threshold adjustment result.

[0047] As a preferred solution, the non-crossing module further comprises a distance subunit and a triggering subunit.

[0048] The distance subunit is configured to calculate the nearest angular distance and the farthest angular distance between the vehicle camera perception box and the pedestrian according to a preset bias formula set, wherein the bias formula set comprises a nearest distance calculation formula and a farthest distance calculation formula.

[0049] The triggering subunit is configured to trigger automatic emergency braking of the vehicle if the nearest angular distance and the farthest angular distance do not satisfy a preset threshold value, and the walking state of the pedestrian at the current time and the future predicted time does not satisfy the non-crossing scenario adjustment result.

[0050] As a preferred solution, the nearest distance calculation formula is specifically:

[0051] closeoffset = closeoffset1 + 0.5 * obj_width

[0052] Wherein, closeoffset is the nearest angular distance between the adjusted camera perception box and the pedestrian, closeoffset1 is the nearest angular distance between the camera perception box before adjustment and the pedestrian, and obj_width is the target width of the pedestrian.

[0053] As a preferred solution, the crossing module comprises a confidence unit and an identification unit.

[0054] The confidence unit is configured to set the vehicle brake confidence of the vehicle to be untrusted if the longitudinal velocity of the pedestrian is greater than the lateral velocity of the pedestrian, or the lateral velocity of the pedestrian is less than or equal to a preset value, to obtain a confidence setting result.

[0055] The identification unit is configured to set the pedestrian identification time of the vehicle under the condition of the confidence setting result, to obtain the crossing scenario adjustment result.

[0056] As a preferred solution, the curve module further comprises a low-speed unit, an overspeed unit, a medium-speed unit and a triggering unit.

[0057] The low-speed unit is configured to calculate the curvature value of the vehicle trajectory according to the steering wheel angle of the current vehicle if the vehicle speed of the current vehicle is less than or equal to a first preset value.

[0058] the super-speed unit is configured to calculate the curvature value of the vehicle trajectory according to the yaw rate of the current vehicle if the vehicle speed of the current vehicle is greater than or equal to a second preset value;

[0059] the medium-speed unit is configured to calculate the curvature value of the current vehicle trajectory according to the steering wheel angle and the yaw rate of the current vehicle if the vehicle speed of the current vehicle is greater than the first preset value and less than the second preset value;

[0060] the triggering unit is configured to trigger automatic emergency braking of the current vehicle if the curvature value of the current vehicle trajectory is greater than the preset value.

[0061] The application further provides a storage medium, wherein the storage medium stores a computer program, the computer program is invoked and executed by a computer, and a complex-scene vehicle adaptive braking optimization method is realized. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 is a flowchart of a complex-scene vehicle adaptive braking optimization method provided by an embodiment of the application;

[0063] Figure 2 is a non-crossing scene diagram provided by an embodiment of the application;

[0064] Figure 3 is a control flowchart provided by an embodiment of the application;

[0065] Figure 4 is a structural diagram of a complex-scene vehicle adaptive braking optimization device provided by an embodiment of the application. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, but not all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0067] In the description of the application, it should be understood that the terms "first" and "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first" and "second" can explicitly or implicitly include one or more of the features. In the description of the application, unless otherwise specified, the meaning of "several" is two or more.

[0068] It should be noted that in the description of the present application, FCW refers to a forward collision warning system (Forward Collision Warning, FCW for short), and AEB refers to an autonomous emergency braking system (Autonomous Emergency Braking, AEB for short).

[0069] The vehicle adaptive braking optimization method for a complex scene provided by the embodiment of the present application is mainly applied to accurately judge the longitudinal distance and speed of a target object, thereby reducing the probability of false triggering of the FCW alarm or AEB function of the vehicle in a complex scene, and maintaining the normal driving condition of the driver.

[0070] Embodiment one:

[0071] Please refer to Figure 1 The embodiment of the present application provides a vehicle adaptive braking optimization method for a complex scene, which comprises S1-S4, and the specific implementation steps are as follows:

[0072] S1, in the case that the pedestrian is in a non-crossing scene, increase the bias value threshold of the pedestrian invading the driving path of the vehicle to obtain a non-crossing scene adjustment result.

[0073] The step S1 of the embodiment of the present application comprises S1.1-S1.3, which are specifically:

[0074] S1.1, obtaining comprehensive information by acquiring external information and self-vehicle information of the vehicle through a self-driving sensor of the vehicle;

[0075] The external information includes target vehicle type, target vehicle transverse and longitudinal distance, target vehicle speed, target vehicle acceleration, lane line width C0, lane line longitudinal distance and lane line curvature C1, and the self-vehicle information includes YawRate (yaw rate), self-vehicle speed, steering wheel angle, self-vehicle acceleration and self-vehicle acceleration.

[0076] S1.2, based on the case that the pedestrian is in a non-crossing scene, according to the comprehensive information, in a one-dimensional condition, the lateral speed of the pedestrian is set to be less than a preset value; in a two-dimensional condition, the longitudinal speed of the pedestrian is set to be less than a preset value; in a three-dimensional condition, the vehicle speed of the vehicle is set to be less than a preset value, to obtain a bias value threshold adjustment result;

[0077] The threshold time age=N for the existence of the pedestrian target is increased by 0.2s to obtain a confirmation frame adjustment result;

[0078] The non-crossing scene adjustment result is composed of the bias value threshold adjustment result and the confirmation frame adjustment result.

[0079] The specific adjustment logic is: a column of content is added in the vlat (pedestrian lateral velocity) and vlgt (pedestrian longitudinal velocity) two-dimensional lookup table of the vehicle, which is "[-PL (lateral offset distance, - represents reduction), 0, 0, 0, 0, 0, 0, 0, 0, 0], PL (lateral offset distance, - represents reduction)"; and the one-dimensional condition is set as: the lateral velocity is vlat, which is less than a certain value, such as 0.8 m / s; the two-dimensional condition is set as: the longitudinal velocity vlgt is 0->30 m / s; the three-dimensional condition is set as: the vehicle speed is V, which is less than a certain value, such as 3.5 m / s; after modification, the current position and predicted position of the pedestrian at the triggering time are no longer the in-path target, and the AEB is no longer triggered; at the same time, the confirmation frame of the pedestrian existence threshold time age=N is increased by 0.2 s; wherein the units of vlat and vlgt are m / s.

[0080] The embodiment S1.2 can accurately distinguish whether the pedestrian is in a stationary, slow walking or ready to cross the road state by setting the lateral velocity of the pedestrian less than a preset value under one-dimensional condition, which helps the system to identify and respond in advance to avoid potential collision risks. Under two-dimensional condition, considering the longitudinal velocity of the pedestrian, the behavior pattern of the pedestrian can be further refined, such as distinguishing whether the pedestrian is accelerating, decelerating or walking at a constant speed, so as to more accurately judge its influence on the vehicle driving path. Under three-dimensional condition, the vehicle speed is set to be less than a preset value, which can ensure the vehicle to maintain a safe speed when driving near the pedestrian, reducing the collision risk caused by too high speed; therefore, by adjusting the bias value threshold in different dimensions, the driving path of the vehicle can be more accurately controlled, so that it can make more delicate and accurate avoidance actions when encountering pedestrians;

[0081] In addition, increasing the confirmation frame of the pedestrian existence threshold time can make the vehicle detect the existence of the pedestrian at an earlier time and make corresponding early warning and response.

[0082] S1.3, calculate the closest angular distance closeoffset and the farthest angular distance faroffset of the vehicle camera perception box from the pedestrian according to the preset bias formula set; wherein the bias formula set includes a nearest distance calculation formula and a farthest distance calculation formula, and the closest angular distance closeoffset and the farthest angular distance faroffset are combined into a square perception box of the vehicle;

[0083] If the closest angular distance closeoffset and the farthest angular distance faroffset do not meet the preset threshold, and the walking state of the pedestrian at the current time and the future predicted time (TTR) does not meet the non-crossing scene adjustment result, the collision warning and automatic emergency braking of the current vehicle are triggered.

[0084] For example, at the triggering moment, the target closeoffset is reduced to 0.25 m, and the faroffset is also shortened by half, so that the pedestrian is no longer considered to be in the self-vehicle road; after the lateral threshold judgment of the target vlat is less than 0.8 m / s, the pedestrian needs to be in the self-vehicle lane at the current moment and the TTR moment to meet the inpath (in the driving path of the ego vehicle) requirement, so that the current pedestrian is no longer considered to be inpath, and the FCW and AEB are no longer triggered.

[0085] The closest distance calculation formula is:

[0086] closeoffset = closeoffset1 + 0.5 * obj_width

[0087] The farthest distance calculation formula is:

[0088] faroffset = faroffset1 - 0.5 * obj_width

[0089] Wherein, closeoffset is the closest angular distance between the camera perception frame of the adjusted vehicle and the pedestrian, closeoffset1 is the closest angular distance between the camera perception frame of the vehicle before adjustment and the pedestrian, obj_width is the target width of the pedestrian, faroffset is the farthest angular distance between the camera perception frame of the adjusted vehicle and the pedestrian, and faroffset1 is the farthest angular distance between the camera perception frame of the vehicle before adjustment and the pedestrian.

[0090] To apply the embodiments of the present application, please refer to Figure 2 , Figure 2 The non-crossing scene schematic diagram provided by the embodiments of the present application shows the target pedestrian or target motor vehicle, which is a non-crossing scene for the current vehicle itself.

[0091] It should be noted that when the pedestrian is in the non-crossing scene, the pedestrian does not cross the front of the vehicle laterally. They may walk, stand or be stationary along the road parallel to the driving direction of the vehicle; at this time, the relative position relationship between the pedestrian and the vehicle is relatively stable, and there is no direct collision risk (unless the pedestrian suddenly changes direction or speed).

[0092] The embodiments S1.3 can accurately calculate the closest angular distance and the farthest angular distance between the camera perception frame of the vehicle and the pedestrian through the preset bias formula set, which helps the system to grasp the position information of the pedestrian in real time, and provides accurate basis for subsequent avoidance and braking decisions. Through the preset bias formula set and threshold judgment, the system can quickly make decisions whether to trigger collision warning and automatic emergency braking in a short time, and this efficient decision mechanism helps to reduce the response time of the system and improve the overall response efficiency.

[0093] S2, filtering the lateral velocity of the pedestrian in the case that the pedestrian is in a crossing scenario, to obtain a crossing scenario adjustment result.

[0094] The step S2 in the embodiments of the present application is specifically:

[0095] In the case that the pedestrian is in a crossing scenario, according to the comprehensive information, if the longitudinal velocity vlgt of the pedestrian is greater than the lateral velocity vlat of the pedestrian, or the lateral velocity vlat of the pedestrian is less than or equal to a preset value, the vehicle brake confidence of the vehicle is set as untrusted, to obtain a confidence setting result.

[0096] Under the condition of the confidence setting result, the pedestrian recognition time of the vehicle is set, and the lateral bias compensation is increased in the manner of increasing the inpathoffset threshold value as in the non-crossing pedestrian scenario, to obtain a crossing scenario adjustment result.

[0097] The specific adjustment logic is that: the target brake confidence of the vehicle is set as untrusted when (target vlgt longitudinal velocity > vlat lateral velocity) || (vlat lateral velocity <= 1.38 m / s) and age < 50, the lateral velocity is adjusted to 0 for the pedestrian and two-wheeled vehicle targets recognized as off-road pedestrians, and the target age is less than 50, and the inpath dangerous target is no longer judged (the target is no longer considered as inpath at the current time and the predicted target position prediction).

[0098] If the driving state of the pedestrian does not meet the crossing scenario adjustment result, the collision warning and automatic emergency braking of the current vehicle are triggered.

[0099] It should be noted that the case that the pedestrian is in a crossing scenario includes three possible pedestrian behaviors or states: the pedestrian is crossing or intending to cross the road; the pedestrian is partially invisible due to the obstruction of roadside objects (such as trees, buildings, etc.); the pedestrian has a certain lateral velocity although mainly moving longitudinally (i.e. along the road direction), and has not yet entered the driving path of the autonomous vehicle.

[0100] In the embodiment S2, when the longitudinal velocity of the pedestrian is greater than the lateral velocity, it may mean that the pedestrian is straight ahead rather than crossing the road; at this time, if the system mistakenly identifies the pedestrian as a crossing scenario and triggers emergency braking, it may cause unnecessary danger and accidents; therefore, setting the brake confidence as untrusted can avoid such misjudgment, thereby improving the driving safety. Moreover, if the lateral velocity of the pedestrian is less than or equal to a preset value, it may indicate that the pedestrian is in a stationary or slow-moving state, in which case setting the brake confidence as untrusted can reduce the emergency braking triggered by the slight movement of the pedestrian, avoiding unnecessary fright and discomfort to the driver and passengers.

[0101] S3. When pedestrians are in a curved scene, reduce the curvature value of the vehicle trajectory to a preset value to obtain the curved scene adjustment result.

[0102] Step S3 in this embodiment of the application is specifically as follows:

[0103] In scenarios where pedestrians are on curves, based on comprehensive information, the threshold for judging large curvature scenarios in vehicle trajectories is raised from C. 01 Reduced to C 02 The results of the curve scene adjustment are obtained.

[0104] If the current vehicle speed V is less than or equal to the first preset value V1, the curvature value C0 of the vehicle trajectory is calculated based on the current steering wheel angle.

[0105] If the current vehicle speed V is greater than or equal to the second preset value V2, the curvature value C0 of the vehicle trajectory is calculated based on the current vehicle yaw rate.

[0106] If the current vehicle speed V is greater than the first preset value V1 and less than the second preset value V2, the curvature value C0 of the current vehicle trajectory is calculated based on the weighted average of the steering wheel angle and yaw rate of the current vehicle.

[0107] If the curvature value C0 of the current vehicle trajectory is greater than C 02 If this occurs, it will trigger a collision warning and automatic emergency braking for the current vehicle.

[0108] It should be noted that the threshold for judging large curvature scenes is changed from C 01 Reduced to C 02 This data change transforms most low-curvature scenarios into high-curvature scenarios. This change reduces the driver's reaction time in high-curvature scenarios: the original reaction time setting was "0 reaction time" for low curvature and "-T2 reaction time" for high curvature; now it's changed to a table lookup based on curvature [C]. 02 C 01 The reaction time range [-T1, -T2] is determined by this factor; after adjustment, the reaction time is reduced by approximately T3, which means that the system needs to calculate a smaller longitudinal deceleration to avoid collisions; therefore, the longitudinal deceleration conditions for FCW and AEB become more difficult to satisfy, thus suppressing triggering in these scenarios.

[0109] For greater than C 01 In scenarios with curvature, the trigger time is not reduced. However, the system can suppress potential risks in these scenarios by increasing lateral offset compensation, such as increasing the inpathoffset threshold for pedestrians intruding into the vehicle's path in non-pedestrian crossing scenarios.

[0110] The formula for calculating the curvature value is as follows:

[0111]

[0112] Wherein, C0 is the current self-vehicle driving estimated curvature, K1 is the inverse of the fluctuation amplitude of the Yawrate value when the self-vehicle drives along a fixed curvature route, K3 is the inverse of the fluctuation amplitude of the steering wheel angle when the self-vehicle drives along a fixed curvature route; K2 is a filtering coefficient, used to avoid frequent jumping of the curvature; Yawrate is the self-vehicle yaw rate, V is the self-vehicle speed, V1 and V2 are respectively the curvature estimated by using the steering wheel angle and the curvature estimated by using the self-vehicle yaw rate; θ is the steering wheel angle, L is the self-vehicle wheelbase, and i is the steering mechanism speed ratio.

[0113] It should be noted that in the pedestrian-in-curve scenario, the vehicle is driving on a curve, and the pedestrian can be located on one side of the curve or on the road. Due to the existence of the curve, the driving trajectory of the vehicle will change, resulting in a complex relative positional relationship between the vehicle and the pedestrian. In addition, there are limitations in simply relying on the curvature perceived by the camera to judge the size of the curve for target processing. In the case of limited curve recognition distance and the case of frequent jumping of the curve curvature, target recognition may become unstable.

[0114] In the embodiment S3, when the vehicle drives at a low speed, the change of the steering wheel angle has a greater impact on the vehicle trajectory, and therefore the trajectory curvature value calculated by the steering wheel angle can more accurately reflect the actual driving state of the vehicle. When the vehicle drives at a high speed, the change of the yaw rate has a more significant impact on the vehicle trajectory; at this time, the trajectory curvature value calculated by the yaw rate can more accurately predict the driving trajectory of the vehicle, thereby improving the driving safety. When the vehicle drives at a medium speed, the changes of the steering wheel angle and the yaw rate both have an impact on the vehicle trajectory; therefore, comprehensively considering both can more accurately calculate the trajectory curvature value, thereby improving the comprehensiveness and accuracy of the calculation.

[0115] S4, the emergency braking adjustment result of the vehicle is composed of the non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result.

[0116] The step S4 of the embodiment of the present application is specifically:

[0117] The emergency braking adjustment result of the vehicle is composed of the non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result.

[0118] Once it is detected that the potential collision body does not satisfy the crossing behavior of the emergency braking adjustment result, the emergency braking and collision warning of the current vehicle are triggered immediately to maximize the possibility of reducing the collision; and the best avoidance path is planned out rapidly, and the driving speed and direction of the vehicle are adjusted to ensure the safety of the pedestrian and the vehicle passing through.

[0119] For the application of the embodiments of the present application, please refer to Figure 3 , Figure 3 is the control flowchart provided by the embodiments of the present application, which shows the general process of the embodiments of the present application for setting the vehicle in different scenarios. Specifically,

[0120] 1) For the longitudinal or stationary pedestrian scenario (pedestrian in non-crossing scenario), when the lateral speed or distance judgment is not accurate, the pedestrian target lateral distance bias is increased, that is, the misrecognized pedestrian target is pushed out of the driving path; at the same time, the lane line is used as the judgment scenario, and the lateral acceleration is directly filtered to be 0, so as to avoid mis-triggering;

[0121] 2) For the short-time appearing or partially occluded scenario (pedestrian in crossing scenario), the pedestrian lateral speed suppression function is directly filtered out, so as to avoid mis-triggering;

[0122] 3) For the pedestrian in the curve scenario, different threshold values are set for the curve calculation to judge whether the current is a large-curvature curve or a small-curvature curve, the system reaction time in the large-curvature scenario is reduced, and the required longitudinal deceleration for collision avoidance is smaller, and the longitudinal deceleration condition is more difficult to meet, so as to avoid mis-triggering.

[0123] It should be noted that the object triggering the current vehicle collision warning and automatic emergency braking function involved in the present application can cover pedestrians, motor vehicles, animals and other potential collision bodies.

[0124] Overall, the embodiments have the following beneficial effects:

[0125] When the pedestrian is in a non-crossing scenario, if the distance between the pedestrian and the vehicle is too close, but there is no actual collision risk, by increasing the bias value threshold of the pedestrian invading the driving path of the vehicle, the relative position and distance between the pedestrian and the vehicle can be more flexibly considered by the AEB system of the vehicle when judging whether the emergency brake needs to be triggered. In the pedestrian crossing scenario, the actual motion state of the pedestrian can be more accurately identified by filtering the lateral speed information of the pedestrian. When the pedestrian indeed has the risk of crossing the driving path of the vehicle, the AEB system will trigger the emergency brake; when the pedestrian only stays for a short time or walks slowly, the system will not mis-trigger. In the curve scenario, due to the change of the driving trajectory and the front view of the vehicle, the traditional AEB system may misjudge the object on the inside of the curve or the obstacle on the roadside in the curve with no danger as a potential collision risk, thereby triggering the emergency brake. By reducing the curvature value of the vehicle trajectory to a preset value, the AEB system can be more flexible and accurate when judging the curve scenario;

[0126] In summary, the present application proposes multiple pedestrian detection and control methods for emergency braking for common pedestrian AEB&FCW false triggering scenarios, which are used to reduce the false triggering of FCW and AEB in pedestrian crossing, non-crossing, curve and other scenarios, can effectively avoid the false triggering of automatic emergency braking in complex road conditions and improve the effect of driving safety, and therefore can solve the problem of false triggering of collision warning and automatic emergency braking of vehicles in complex scenarios.

[0127] Embodiment two:

[0128] Please refer to Figure 4 The embodiment of the present application provides a vehicle adaptive braking optimization device for complex scenarios, which comprises a non-crossing module 10, a crossing module 20, a curve module 30 and a comprehensive module 40.

[0129] The non-crossing module 10 is used to increase the bias value threshold of the pedestrian invading the driving path of the vehicle when the pedestrian is in a non-crossing scenario, and obtain a non-crossing scenario adjustment result.

[0130] The crossing module 20 is used to filter the lateral speed of the pedestrian when the pedestrian is in a crossing scenario, and obtain a crossing scenario adjustment result.

[0131] The curve module 30 is used to reduce the curvature value of the vehicle trajectory to a preset value when the pedestrian is in a curve scenario, and obtain a curve scenario adjustment result.

[0132] The comprehensive module 40 is used to form an emergency braking adjustment result of the vehicle by the non-crossing scenario adjustment result, the crossing scenario adjustment result and the curve scenario adjustment result.

[0133] In one embodiment, the non-crossing module 10 comprises a data unit, a bias unit, a time unit, a combination unit, a distance subunit and a trigger subunit.

[0134] The data unit is used to obtain comprehensive information by acquiring external information and self-vehicle information of the vehicle through self-driving sensors of the vehicle.

[0135] The external information comprises target vehicle type, target vehicle lateral and longitudinal distance, target vehicle speed, target vehicle acceleration, lane line width C0, lane line longitudinal distance and lane line curvature C1, and the self-vehicle information comprises YawRate (yaw rate), self-vehicle speed, steering wheel angle, self-vehicle acceleration and self-vehicle acceleration.

[0136] The bias unit is used to set the lateral speed of the pedestrian to be less than a preset value in one-dimensional condition, set the longitudinal speed of the pedestrian to be less than a preset value in two-dimensional condition, and set the vehicle speed of the vehicle to be less than a preset value in three-dimensional condition based on the fact that the pedestrian is in a non-crossing scenario according to the comprehensive information, and obtain a bias value threshold adjustment result.

[0137] a time unit for increasing the threshold time age=N of the pedestrian target existence by 0.2s to obtain a confirmation frame adjustment result;

[0138] a combination unit for constructing the non-crossing scene adjustment result from the bias value threshold adjustment result and the confirmation frame adjustment result.

[0139] wherein the specific adjustment logic is: adding a column of contents "[-PL (lateral bias distance, - represents reduction), 0, 0, 0, 0, 0, 0, 0, 0, 0], PL (lateral bias distance, - represents reduction)" in the two-dimensional look-up table of vlat (lateral speed of pedestrian) and vlgt (longitudinal speed of pedestrian) of the vehicle; and the one-dimensional condition is set as: the lateral speed is vlat, and vlat is less than a certain value, such as 0.8 m / s; the two-dimensional condition is set as: the longitudinal speed vlgt is 0->30 m / s; the three-dimensional condition is set as: the vehicle speed is V, and V is less than a certain value, such as 3.5 m / s; the current position and the predicted position of the pedestrian at the triggering moment are no longer in-path targets after modification, and the AEB is no longer triggered; at the same time, the confirmation frame of the threshold time age=N of the pedestrian existence is increased by 0.2s; wherein the units of vlat and vlgt are m / s.

[0140] The bias unit, time unit and combination unit in this embodiment can accurately distinguish whether the pedestrian is in a state of stillness, slow walking or preparing to cross the road by setting the lateral speed of the pedestrian less than a preset value under one-dimensional condition, which helps the system to identify and react in advance and avoid potential collision risks. Under two-dimensional condition, considering the longitudinal speed of the pedestrian, the behavior pattern of the pedestrian can be further refined, such as distinguishing whether the pedestrian is accelerating, decelerating or walking at a constant speed, so as to more accurately judge the influence of the pedestrian on the driving path of the vehicle. Under three-dimensional condition, the vehicle speed of the vehicle is set to be less than a preset value, which can ensure that the vehicle drives at a safe speed when the pedestrian is nearby, reducing the collision risk caused by too high speed; therefore, by adjusting the bias value threshold in different dimensions, the driving path of the vehicle can be more accurately controlled, so that it can make more delicate and accurate avoidance actions when encountering pedestrians;

[0141] Moreover, increasing the confirmation frame of the threshold time of the pedestrian existence can make the vehicle detect the existence of the pedestrian at an earlier time and make corresponding early warning and response.

[0142] The distance subunit is configured to calculate a closest angular distance closeoffset and a farthest angular distance faroffset of a camera perception box of the vehicle from the pedestrian according to a preset bias formula set, wherein the bias formula set comprises a closest distance calculation formula and a farthest distance calculation formula, and the closest angular distance closeoffset and the farthest angular distance faroffset are combined into a square perception box of the vehicle.

[0143] The triggering subunit is configured to trigger a collision warning and automatic emergency braking of the current vehicle if the closest angular distance closeoffset and the farthest angular distance faroffset do not meet a preset threshold, and a walking state of the pedestrian at a current time and a future predicted time (TTR) does not meet a non-crossing scene adjustment result.

[0144] For example, at the triggering time, the target closeoffset is shortened to 0.25 m, and the faroffset is also shortened by half, so the pedestrian is no longer considered to be in the road of the ego vehicle; after the lateral threshold judgment that the target vlat is less than 0.8 m / s, the pedestrian needs to be in the lane of the ego vehicle at the current time and the TTR time to meet the inpath requirement (in the driving path of the ego vehicle), so the current pedestrian is no longer considered to be in the path, and the FCW and AEB are no longer triggered.

[0145] The closest distance calculation formula is as follows:

[0146] closeoffset = closeoffset1 + 0.5 * obj_width

[0147] The farthest distance calculation formula is as follows:

[0148] faroffset = faroffset1 - 0.5 * obj_width

[0149] The closest distance calculation formula is as follows:

[0150] For application of the embodiments of the present application, please refer to Figure 2 , Figure 2 is a non-crossing scene diagram provided by the embodiments of the present application, which shows a target pedestrian or a target motor vehicle, and is a non-crossing scene for the current vehicle itself.

[0151] It should be noted that in the non-crossing scenario of the pedestrian, the pedestrian does not cross the front of the vehicle in the transverse direction. They may walk, stand or be stationary along the road parallel to the driving direction of the vehicle; at this time, the relative position relationship between the pedestrian and the vehicle is relatively stable, and there is no direct collision risk (unless the pedestrian suddenly changes direction or speed).

[0152] The distance subunit and the triggering subunit can accurately calculate the nearest angular distance and the farthest angular distance of the vehicle camera perception box and the pedestrian through the preset bias formula set, which helps the system to master the position information of the pedestrian in real time, and provides accurate basis for subsequent avoidance and braking decisions. Through the preset bias formula set and threshold judgment, the system can quickly make decisions on whether to trigger collision warning and automatic emergency braking in a short time. This efficient decision mechanism helps to reduce the response time of the system and improve the overall response efficiency.

[0153] In one embodiment, the crossing module 20 includes a confidence unit and an identification unit;

[0154] The confidence unit is configured to, in the case that the pedestrian is in a crossing scenario, set the vehicle brake confidence of the vehicle as untrusted according to the comprehensive information if the longitudinal speed vlgt of the pedestrian is greater than the lateral speed vlat of the pedestrian, or the lateral speed vlat of the pedestrian is less than or equal to a preset value, to obtain a confidence setting result.

[0155] The identification unit is configured to, under the condition of the confidence setting result, set the pedestrian identification time of the vehicle, and increase the lateral bias compensation in the manner of increasing the inpathoffset threshold value as in the non-crossing pedestrian scenario, to obtain a crossing scenario adjustment result.

[0156] The specific adjustment logic is: (target vlgt longitudinal speed > vlat lateral speed) || (vlat lateral speed <= 1.38 m / s) and age < 50, the vehicle target brake confidence will be set as untrusted, the target lateral speed is adjusted to 0 for the pedestrian and two-wheeled vehicle identified outside the road, and the target age is less than 50, and the inpath dangerous target is no longer judged (the target is no longer considered as inpath at the current time and the predicted time target position prediction).

[0157] If the driving state of the pedestrian does not meet the crossing scenario adjustment result, the collision warning and automatic emergency braking of the current vehicle is triggered.

[0158] It should be noted that the situation of the pedestrian crossing the scene includes three possible pedestrian behaviors or states: the pedestrian is crossing or intending to cross the road; the pedestrian is partially invisible due to the obstruction of roadside objects (such as trees, buildings, etc.); and the pedestrian has a certain lateral speed although mainly moving longitudinally (i.e., along the direction of the road) and has not yet entered the driving path of the autonomous vehicle.

[0159] In the crossing module 20 of the embodiment, when the longitudinal speed of the pedestrian is greater than the lateral speed, it may mean that the pedestrian is straightening rather than crossing the road; at this time, if the system mistakenly identifies the pedestrian as a crossing scene and triggers emergency braking, it may cause unnecessary danger and accidents; therefore, setting the brake confidence to be untrusted can avoid such misjudgment, thereby improving driving safety. Moreover, if the lateral speed of the pedestrian is less than or equal to a preset value, it may indicate that the pedestrian is in a stationary or slow-moving state; in this case, setting the brake confidence to be untrusted can reduce the triggering of emergency braking due to slight movement of the pedestrian, thereby avoiding unnecessary fright and discomfort to the driver and passengers.

[0160] In one embodiment, the curve module 30 includes an adjusting unit, a low-speed unit, an overspeed unit, a medium-speed unit, and a triggering unit.

[0161] The adjusting unit is configured to, in a case where the pedestrian is in a curve scene, adjust a judgment threshold value of a large-curvature scene in the vehicle trajectory from C 01 to C 02 to obtain a curve scene adjustment result.

[0162] The low-speed unit is configured to, if the speed V of the current vehicle is less than or equal to a first preset value V1, calculate a curvature value C0 of the vehicle trajectory according to the steering wheel angle of the current vehicle.

[0163] The overspeed unit is configured to, if the speed V of the current vehicle is greater than or equal to a second preset value V2, calculate a curvature value C0 of the vehicle trajectory according to the yaw rate of the current vehicle.

[0164] The medium-speed unit is configured to, if the speed V of the current vehicle is greater than the first preset value V1 and less than the second preset value V2, calculate a curvature value C0 of the current vehicle trajectory according to a weighted average of the steering wheel angle and the yaw rate of the current vehicle.

[0165] The triggering unit is configured to, if the curvature value C0 of the current vehicle trajectory is greater than C 02 , trigger a collision warning and automatic emergency braking of the current vehicle.

[0166] It should be noted that the judgment threshold value of the large-curvature scene is lowered from C 01 to C 02, the data change will turn most small-curvature scenarios into large-curvature scenarios. This change reduces the reaction time of the driver in large-curvature scenarios: the original reaction time setting is that small curvature corresponds to "0 reaction time" and large curvature corresponds to "-T2 reaction time", and now it is changed to determine the reaction time range [-T1, -T2] according to the curvature lookup table [C 02 ,C 01 ]; the adjusted reaction time is reduced by about T3, which means that the system needs to calculate a smaller longitudinal deceleration to avoid a collision; therefore, the longitudinal deceleration condition of FCW and AEB becomes more difficult to meet, thereby suppressing the triggering in these scenarios;

[0167] For scenarios greater than C 01 , the triggering time does not decrease. However, the system can suppress the potential risk in these scenarios by increasing the lateral offset compensation, such as increasing the in-path offset threshold value of the non-crossing pedestrian scenario.

[0168] wherein the formula for calculating the curvature value is:

[0169]

[0170] wherein C0 is the current vehicle driving estimated curvature, K1 is the inverse of the fluctuation amplitude of the Yawrate value when the vehicle is driving along a fixed curvature route, K3 is the inverse of the fluctuation amplitude of the steering wheel angle when the vehicle is driving along a fixed curvature route; K2 is a filtering coefficient used to avoid frequent jumps in curvature; Yawrate is the vehicle yaw rate, V is the vehicle speed, V1 and V2 are the curvatures estimated using the steering wheel angle and the vehicle trajectory, respectively; θ is the steering wheel angle, L is the vehicle wheelbase, and i is the steering mechanism speed ratio.

[0171] It should be noted that in the pedestrian-in-curve scenario, the vehicle is driving on a curve, and the pedestrian may be located on one side of the curve or on the road. Due to the existence of the curve, the driving trajectory of the vehicle will change, causing the relative position relationship between the vehicle and the pedestrian to become complex. In addition, simply relying on the curvature perceived by the camera output to determine the size of the curve for target processing has limitations. In the case of limited curve recognition distance and frequent jumps in curve curvature, target recognition may become unstable.

[0172] In the curve module 30 of the embodiment, when the vehicle is running at low speed, the change of the steering wheel angle has a greater impact on the vehicle trajectory, and thus the trajectory curvature value calculated by the steering wheel angle can more accurately reflect the actual driving state of the vehicle. When the vehicle is running at high speed, the change of the yaw rate has a more significant impact on the vehicle trajectory; at this time, the trajectory curvature value calculated by the yaw rate can more accurately predict the driving trajectory of the vehicle, thereby improving the driving safety. When the vehicle is running at medium speed, the changes of the steering wheel angle and the yaw rate both have an impact on the vehicle trajectory; thus, the trajectory curvature value can be more accurately calculated by comprehensively considering both, thereby improving the comprehensiveness and accuracy of the calculation.

[0173] In one embodiment, the comprehensive module 40 specifically comprises:

[0174] The emergency braking adjustment result of the vehicle is composed of the non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result.

[0175] Once the potential collision body crossing behavior that does not meet the emergency braking adjustment result is detected, the emergency braking and collision warning of the current vehicle are triggered immediately to minimize the possibility of collision; and the best avoidance path is planned out quickly, and the driving speed and direction of the vehicle are adjusted to ensure the safety of the pedestrian and the vehicle passing through.

[0176] For the application of the embodiments of the present application, please refer to Figure 3 , Figure 3 is a control flowchart provided by the embodiments of the present application, which shows the general process of the second embodiment of the present application for making corresponding settings for the vehicle in different scenes, specifically comprising:

[0177] 1) For the longitudinal or stationary pedestrian scene (the pedestrian is in the non-crossing scene), when the lateral speed or distance judgment is not accurate, the pedestrian target lateral distance bias is increased, that is, the misrecognized pedestrian target is pushed out of the driving path; at the same time, the lane line is used as the judgment scene to directly filter the lateral acceleration of 0, thereby avoiding mis-triggering;

[0178] 2) For the scene with short appearance time or partial occlusion (the pedestrian is in the crossing scene), the pedestrian lateral speed suppression function triggering is directly filtered out, thereby avoiding mis-triggering;

[0179] 3) For the pedestrian in the curve scene, different thresholds are set for the curve curvature calculation to judge whether the current is a large curvature curve or a small curvature curve, the system reaction time in the large curvature scene is reduced, and the smaller the longitudinal deceleration required for collision avoidance is calculated, the more difficult the longitudinal deceleration condition is to meet, thereby avoiding mis-triggering.

[0180] It should be noted that the object triggering the current vehicle collision warning and automatic emergency braking function involved in the present application can cover pedestrians, motor vehicles and animals and other potential collision bodies.

[0181] Overall, the embodiment has the following beneficial effects:

[0182] In the non-crossing scenario of the pedestrian, if the distance between the pedestrian and the vehicle is too close but there is no actual collision risk, by increasing the bias value threshold of the pedestrian intrusion into the driving path of the vehicle, the AEB system of the vehicle can more flexibly consider the relative position and distance between the pedestrian and the vehicle when judging whether to trigger emergency braking. In the crossing scenario of the pedestrian, by filtering the lateral speed information of the pedestrian, the actual motion state of the pedestrian can be more accurately identified, and when the pedestrian does indeed have the risk of crossing the driving path of the vehicle, the AEB system will trigger emergency braking; when the pedestrian only stays for a short time or walks slowly, the system will not be triggered. In the curve scenario, due to the change of the driving trajectory and the front view of the vehicle, the traditional AEB system may misjudge the object inside the curve or the obstacle on the roadside in front of the vehicle without danger as a potential collision risk, thereby triggering emergency braking; by reducing the curvature value of the vehicle trajectory to a preset value, the AEB system can be more flexible and accurate when judging the curve scenario;

[0183] In summary, the present application proposes various emergency braking pedestrian detection and control methods for common pedestrian AEB&FCW mis-triggering scenarios, which can reduce the mis-triggering of FCW and AEB in pedestrian crossing, non-crossing and curve scenarios, effectively avoid the mis-triggering of automatic emergency braking in complex road conditions and improve driving safety, and therefore can solve the problem of easy mis-triggering of collision warning and automatic emergency braking of vehicles in complex scenarios.

[0184] Embodiment three:

[0185] The embodiment of the present application provides a computer readable storage medium, which comprises a stored computer program, wherein the computer program controls a device where the computer readable storage medium is located to execute the adaptive braking optimization method for vehicles in complex scenarios when the computer program runs.

[0186] The vehicle adaptive braking optimization method for a complex scene can be stored in a computer readable storage medium if it is implemented in the form of a software function unit and used as an independent product. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each method embodiment described above can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, an executable file, or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, a recording medium, a U disk, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0187] The above is the preferred embodiment of the present application. It should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which are also considered within the scope of protection of the present application.

Claims

1. A method for vehicle adaptive braking optimization for complex scenarios, characterized in that, Comprise: In the case of a pedestrian in a non-crossing scene, increase the bias value threshold of the pedestrian invading the driving path of the vehicle to obtain a non-crossing scene adjustment result; Specifically: based on the case that the pedestrian is in a non-crossing scene, in one-dimensional conditions, the lateral speed of the pedestrian is set to be less than a preset value; in two-dimensional conditions, the longitudinal speed of the pedestrian is set to be less than a preset value; in three-dimensional conditions, the vehicle speed of the vehicle is set to be less than a preset value, to obtain the bias value threshold adjustment result; increase the confirmation frame of the pedestrian existing threshold time to obtain the confirmation frame adjustment result; the bias value threshold adjustment result and the confirmation frame adjustment result constitute the non-crossing scene adjustment result; In the case of the pedestrian in a crossing scene, filter the lateral speed of the pedestrian to obtain a crossing scene adjustment result; In the case of the pedestrian in a curve scene, reduce the curvature value of the vehicle trajectory to a preset value to obtain a curve scene adjustment result; The non-crossing scene adjustment result, the crossing scene adjustment result and the curve scene adjustment result constitute the emergency braking adjustment result of the vehicle.

2. The vehicle adaptive braking optimization method for complex scenarios of claim 1, wherein, After obtaining the non-crossing scene adjustment result, further comprising: According to the preset bias formula set, the closest angular distance and the farthest angular distance between the vehicle camera perception box and the pedestrian are calculated; wherein the bias formula set includes a nearest distance calculation formula and a farthest distance calculation formula; If the closest angular distance and the farthest angular distance do not meet the preset threshold, and the walking state of the pedestrian at the current time and the future predicted time does not meet the non-crossing scene adjustment result, the automatic emergency braking of the vehicle is triggered.

3. The method for vehicle adaptive braking optimization of a complex scenario according to claim 2, wherein, The nearest distance calculation formula is specifically: closeoffset=closeoffset1+0.5*obj_width Wherein, closeoffset is the closest angular distance between the adjusted camera perception box and the pedestrian, closeoffset1 is the closest angular distance between the camera perception box and the pedestrian before adjustment, and obj_width is the pedestrian target width.

4. The method for vehicle adaptive braking optimization for complex scenarios as claimed in claim 1 wherein, In the case of the pedestrian in a crossing scene, filter the lateral speed of the pedestrian to obtain a crossing scene adjustment result, specifically: If the longitudinal speed of the pedestrian is greater than its lateral speed, or the lateral speed of the pedestrian is less than or equal to a preset value, set the vehicle brake confidence of the vehicle to be untrusted to obtain a confidence setting result; Under the condition of the confidence setting result, set the pedestrian recognition time of the vehicle to obtain the crossing scene adjustment result.

5. The method for vehicle adaptive braking optimization for complex scenarios as claimed in claim 1 wherein, After obtaining the curve scene adjustment result, further comprising: If the current vehicle speed is less than or equal to a first preset value, the curvature value of the vehicle trajectory is calculated according to the steering wheel angle of the current vehicle; If the current vehicle speed is greater than or equal to a second preset value, the curvature value of the vehicle trajectory is calculated according to the yaw rate of the current vehicle; If the current vehicle speed is greater than the first preset value and less than the second preset value, the curvature value of the current vehicle trajectory is calculated according to the steering wheel angle and the yaw rate of the current vehicle; If the curvature value of the current vehicle trajectory is greater than the preset value, the automatic emergency braking of the current vehicle is triggered.

6. A vehicle adaptive brake optimization device for complex scenarios, characterized by, The non-crossing module, the crossing module, the curve module, and the comprehensive module are included. The non-crossing module is configured to increase a bias value threshold of a pedestrian invading a driving path of a vehicle in a non-crossing scenario to obtain a non-crossing scenario adjustment result. Specifically, the lateral speed of the pedestrian is set to be less than a preset value in a one-dimensional condition, the longitudinal speed of the pedestrian is set to be less than a preset value in a two-dimensional condition, and the vehicle speed of the vehicle is set to be less than a preset value in a three-dimensional condition, to obtain a bias value threshold adjustment result. The confirmation frame of a pedestrian existing threshold time is increased to obtain a confirmation frame adjustment result. The bias value threshold adjustment result and the confirmation frame adjustment result constitute the non-crossing scenario adjustment result. The crossing module is configured to filter the lateral speed of the pedestrian in a crossing scenario to obtain a crossing scenario adjustment result. The curve module is configured to reduce the curvature value of the vehicle trajectory to a preset value in a curve scenario to obtain a curve scenario adjustment result. The comprehensive module is configured to constitute an emergency braking adjustment result of the vehicle from the non-crossing scenario adjustment result, the crossing scenario adjustment result, and the curve scenario adjustment result.

7. A vehicle adaptive brake optimization device for a complex scenario according to claim 6, wherein The non-crossing module further includes a distance subunit and a triggering subunit. The distance subunit is configured to calculate the nearest angular distance and the farthest angular distance between the vehicle camera perception frame and the pedestrian according to a preset bias formula set. The bias formula set includes a nearest distance calculation formula and a farthest distance calculation formula. The triggering subunit is configured to trigger the automatic emergency braking of the vehicle if the nearest angular distance and the farthest angular distance do not satisfy a preset threshold value, and the walking state of the pedestrian at the current time and the future predicted time does not satisfy the non-crossing scenario adjustment result.

8. A vehicle adaptive brake optimization device for a complex scenario according to claim 7, wherein, The nearest distance calculation formula is specifically: closeoffset=closeoffset1+0.5*obj_width wherein closeoffset is the adjusted nearest angular distance between the camera perception frame and the pedestrian, closeoffset1 is the nearest angular distance between the camera perception frame and the pedestrian before adjustment, and obj_width is the pedestrian target width.

9. The vehicle adaptive brake optimization device for complex scenarios of claim 6, wherein, The crossing module includes a confidence unit and an identification unit. The confidence unit is configured to set the vehicle brake confidence of the vehicle as untrusted if the longitudinal speed of the pedestrian is greater than the lateral speed of the pedestrian, or the lateral speed of the pedestrian is less than or equal to a preset value, to obtain a confidence setting result. The identification unit is configured to set the pedestrian identification time of the vehicle under the condition of the confidence setting result to obtain the crossing scenario adjustment result.

10. The vehicle adaptive brake optimization device for complex scenarios of claim 6, wherein, The curve module further includes a low-speed unit, an overspeed unit, a medium-speed unit, and a triggering unit. The low-speed unit is configured to calculate the curvature value of the vehicle trajectory according to the steering wheel angle of the current vehicle if the vehicle speed of the current vehicle is less than or equal to a first preset value. The overspeed unit is configured to calculate the curvature value of the vehicle trajectory according to the yaw rate of the current vehicle if the vehicle speed of the current vehicle is greater than or equal to a second preset value. The medium-speed unit is configured to calculate the curvature value of the current vehicle trajectory according to the steering wheel angle and the yaw rate of the current vehicle if the vehicle speed of the current vehicle is greater than the first preset value and less than the second preset value. The triggering unit is configured to trigger the automatic emergency braking of the current vehicle if the curvature value of the current vehicle trajectory is greater than the preset value.

11. A storage medium, characterized by The storage medium has a computer program stored thereon, the computer program is invoked and executed by a computer, and a complex scene vehicle adaptive braking optimization method according to any one of claims 1 to 5 is implemented.

Images